Functions — github.com/acaelles97/DeVIS

hub / github.com/acaelles97/DeVIS / functions

Functions681 in github.com/acaelles97/DeVIS

⨍ Functions 681◇ Types & classes 126

↓ 2 callersMethoddecompose

(self)

src/util/misc.py:369

↓ 2 callersMethodevaluate

Evaluate a set of metrics on a set of datasets

src/trackeval/eval.py:49

↓ 2 callersFunctionevaluate_coco

(model: torch.nn.Module, criterion: torch.nn.Module, postprocessors, data_loader: DataLoader

src/engine.py:99

↓ 2 callersMethodforward_ffn

(self, src)

src/models/deformable_transformer.py:159

↓ 2 callersFunctionget_args_parser

()

main.py:27

↓ 2 callersMethodget_class_name

(cls)

src/trackeval/datasets/_base_dataset.py:47

↓ 2 callersMethodget_default_metric_config

Default class config values

src/trackeval/metrics/track_map.py:13

↓ 2 callersMethodget_loss

(self, loss, outputs, targets, indices, num_boxes, **kwargs)

src/models/criterion.py:199

↓ 2 callersMethodget_mask_id

(self)

src/models/tracker.py:71

↓ 2 callersFunctionget_most_left_coordinate

(mask)

src/util/viz_utils.py:12

↓ 2 callersMethodget_reference_points

(spatial_shapes, valid_ratios, device)

src/models/deformable_transformer.py:185

↓ 2 callersMethodget_src_permutation_idx

(indices: List[Tensor])

src/models/deformable_segmentation.py:151

↓ 2 callersFunctionimshow_det_bboxes_for_tracks

(img, instance, cmap, class_names=None, thickness=2,

src/util/viz_utils.py:30

↓ 2 callersFunctioninterpolate

Equivalent to nn.functional.interpolate, but with support for empty batch sizes. This will eventually be supported natively by PyTorch, and t

src/util/misc.py:482

↓ 2 callersFunctionis_main_process

()

src/util/misc.py:421

↓ 2 callersMethodkill_track

(self)

src/models/tracker.py:98

↓ 2 callersMethodlog_every

(self, iterable, epoch=None, header=None)

src/util/misc.py:215

↓ 2 callersFunctionmain

(args, cfg)

main.py:97

↓ 2 callersFunctionmake_train_vis_transforms

(out_scale, multi_scale_training, create_bbx_from_mask)

src/datasets/vis.py:223

↓ 2 callersFunctionmasks_to_boxes

Compute the bounding boxes around the provided masks The masks should be in format [N, H, W] where N is the number of masks, (H, W) are the spati

src/util/box_ops.py:101

↓ 2 callersFunctionms_deform_attn_core_pytorch

(value, value_spatial_shapes, sampling_locations, attention_weights)

src/models/ops/functions/ms_deform_attn_func.py:102

↓ 2 callersFunctionnested_dict_to_namespace

(dictionary)

src/util/misc.py:492

↓ 2 callersMethodprepare_data

(self, srcs, masks, pos_embeds)

src/models/deformable_transformer.py:69

↓ 2 callersMethodprint_table

Prints table of results for all sequences

src/trackeval/metrics/_base_metric.py:66

↓ 2 callersMethodprocess_output

(self, outputs)

src/models/deformable_detr.py:243

↓ 2 callersFunctionread_image_for_axis

(track, t, folder_images, images_path, cmap)

src/util/att_maps_viz.py:115

↓ 2 callersFunctionrowify

(d)

src/trackeval/datasets/run_rob_mots.py:101

↓ 2 callersFunctionsanity_check

(cfg)

main.py:52

↓ 2 callersFunctionshift_class_neurons

(checkpoint: dict)

src/util/weights_loading_utils.py:10

↓ 2 callersFunctionsigmoid_focal_loss

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

src/models/deformable_segmentation.py:401

↓ 2 callersFunctionsort_by_score

Sorts data by score

src/trackeval/baselines/baseline_utils.py:115

↓ 2 callersMethodsynchronize_between_processes

(self, device=None)

src/util/misc.py:199

↓ 2 callersMethodtrain

Convert the model into training mode while keep layers freezed.

src/models/swin_backbone.py:605

↓ 2 callersMethodupdate

(self, value, n=1)

src/util/misc.py:36

↓ 2 callersMethodupdate_stride

(self, stride)

src/models/tracker.py:126

↓ 2 callersMethodupdate_stride_and_encode_masks

(self, stride, overlap_window)

src/models/tracker.py:129

↓ 2 callersFunctionwindow_partition

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

src/models/swin_backbone.py:36

↓ 2 callersMethodwith_pos_embed

(tensor, pos)

src/models/deformable_transformer.py:244

↓ 1 callersMethod__init__

(self, defdetr: DeformableDETR, matcher: DeVISHungarianMatcher, mask_head_used_features: List[List[str]],

src/models/devis_segmentation.py:15

↓ 1 callersMethod__init__

Creates the matcher Params: cost_class: This is the relative weight of the classification error in the matching cost

src/models/matcher.py:132

↓ 1 callersMethod__init__

(self, d_model=256, nhead=8, num_encoder_layers=6, num_decoder_layers=6, dim_feedforward=1024

src/models/devis_ablation_transformer_wo_t_conn.py:18

↓ 1 callersMethod_calculate_similarities

(self, gt_dets_t, tracker_dets_t)

src/trackeval/datasets/_base_dataset.py:41

↓ 1 callersMethod_check_input

(self, k, mask)

src/models/deformable_segmentation.py:296

↓ 1 callersMethod_compute_centroid

(box)

src/trackeval/metrics/ideucl.py:120

↓ 1 callersMethod_compute_image_to_timestep_mappings

Computes a mapping from images to the corresponding timestep in the sequence. :return: the image-to-timestep-mapping

src/trackeval/datasets/tao.py:486

↓ 1 callersMethod_compute_j

Computation of J value for all ground truth IDs and all tracker IDs in the given sequence. Adapted from https://github.com/davisvideo

src/trackeval/metrics/j_and_f.py:275

↓ 1 callersMethod_compute_track_ious

Calculate track IoUs for a set of ground truth tracks and a set of detected tracks

src/trackeval/metrics/track_map.py:434

↓ 1 callersMethod_detailed_row

(self, res)

src/trackeval/metrics/_base_metric.py:125

↓ 1 callersMethod_fill_video_ids_inplace

Fills in missing video IDs inplace. Adapted from https://github.com/TAO-Dataset/ :param annotations: the annotations for which the vi

src/trackeval/datasets/tao.py:523

↓ 1 callersFunction_get_boundaries

(x_values, y_values, round_val)

src/trackeval/plotting.py:140

↓ 1 callersMethod_get_eval_embeddings

(self, out, targets, hs)

src/models/devis_segmentation.py:47

↓ 1 callersMethod_get_eval_embeddings

(self, out, hs)

src/models/deformable_segmentation.py:192

↓ 1 callersMethod_get_features_for_mask_head

(self, backbone_feats: List[Tensor], srcs: List[Tensor], memories: List[Tensor])

src/models/deformable_segmentation.py:92

↓ 1 callersMethod_get_matched_with_filled_embeddings

(self, indices: List[Tensor], hs: Tensor, lvl: int)

src/models/deformable_segmentation.py:174

↓ 1 callersMethod_get_seq_info

(self)

src/trackeval/datasets/rob_mots.py:132

↓ 1 callersMethod_get_seq_info

(self)

src/trackeval/datasets/mot_challenge_2d_box.py:131

↓ 1 callersMethod_get_seq_info

(self)

src/trackeval/datasets/kitti_mots.py:114

↓ 1 callersMethod_get_seq_info

(self)

src/trackeval/datasets/head_tracking_challenge.py:129

↓ 1 callersMethod_get_seq_info

(self)

src/trackeval/datasets/mots_challenge.py:124

↓ 1 callersMethod_get_tgt_permutation_masked_idx

(self, indices)

src/models/criterion.py:79

↓ 1 callersMethod_get_tracker_seq_tracks

Prepares tracker data for a given sequence. Extracts all annotations for given sequence ID, computes average track area and assigns a

src/trackeval/datasets/youtube_vis.py:355

↓ 1 callersMethod_get_training_embeddings

(self, out, targets, hs, lvl)

src/models/devis_segmentation.py:37

↓ 1 callersMethod_get_training_embeddings

(self, out, targets, hs, loss_lvl)

src/models/deformable_segmentation.py:182

↓ 1 callersMethod_inference_forward

(self, targets, out, hs, memories_att_map, masks_att_map, mask_head_feats, spatial_shapes)

src/models/devis_segmentation.py:75

↓ 1 callersMethod_inference_forward

(self, out, hs, memories_att_map, masks_att_map, mask_head_feats)

src/models/deformable_segmentation.py:220

↓ 1 callersMethod_limit_dets_per_image

Limits the number of detections for each image to config['MAX_DETECTIONS']. Adapted from https://github.com/TAO-Dataset/ :par

src/trackeval/datasets/tao.py:503

↓ 1 callersMethod_load_annotations

(self)

src/datasets/vis.py:148

↓ 1 callersMethod_make_track_ids_unique

Makes the track IDs unqiue over the whole annotation set. Adapted from https://github.com/TAO-Dataset/ :param annotations: the annota

src/trackeval/datasets/tao.py:538

↓ 1 callersFunction_max_by_axis

(the_list)

src/util/misc.py:321

↓ 1 callersFunction_plot_bg_contour

Plot background contour.

src/trackeval/plotting.py:178

↓ 1 callersFunction_plot_pareto_optimal_lines

Plot pareto optimal lines

src/trackeval/plotting.py:204

↓ 1 callersMethod_prepare_gt_annotations

Prepares GT data by rle encoding segmentations and computing the average track area. :return: None

src/trackeval/datasets/youtube_vis.py:335

↓ 1 callersMethod_raise_index_error

Auxiliary method to raise an evaluation error in case of an index error while reading files. :param is_gt: whether gt or tracker data

src/trackeval/datasets/rob_mots.py:265

↓ 1 callersMethod_raise_value_error

Auxiliary method to raise an evaluation error in case of an value error while reading files. :param is_gt: whether gt or tracker data

src/trackeval/datasets/rob_mots.py:283

↓ 1 callersMethod_reset_parameters

(self)

src/models/deformable_transformer.py:48

↓ 1 callersMethod_reset_parameters

(self)

src/models/position_encoding.py:117

↓ 1 callersMethod_reset_parameters

(self)

src/models/ops/modules/ms_deform_attn.py:184

↓ 1 callersMethod_sanity_check

(self)

src/models/deformable_segmentation.py:59

↓ 1 callersMethod_set_aux_loss

(self, outputs_class, outputs_coord)

src/models/deformable_detr.py:210

↓ 1 callersMethod_training_forward

(self, targets, out, hs, loss_lvl, memories_att_map, masks_att_map, mask_head_feats)

src/models/devis_segmentation.py:69

↓ 1 callersMethodaccumulate

(self)

src/datasets/coco_eval.py:62

↓ 1 callersFunctionaccuracy

Computes the precision@k for the specified values of k

src/util/misc.py:464

↓ 1 callersFunctionadapt_weights_devis

(checkpoint: dict, model_state_dict: dict, lvl_res: int, focal_loss: bool, finetune_class_logits: bool, num_fr

src/util/weights_loading_utils.py:52

↓ 1 callersMethodadd_video_track

(self, t_window, track)

src/models/tracker.py:86

↓ 1 callersMethodapply_random_sequence_shuffle

(self, images, instance_masks)

src/datasets/coco_joint_vis.py:61

↓ 1 callersMethodbackward

(ctx, grad_output)

src/models/ops/functions/ms_deform_attn_func.py:32

↓ 1 callersFunctionbg_format

(val)

src/trackeval/plotting.py:196

↓ 1 callersFunctionbox_iou

(boxes1, boxes2)

src/util/box_ops.py:61

↓ 1 callersFunctionbuild_3D_conv_head

(mask_head_hidden_size)

src/models/devis_segmentation.py:167

↓ 1 callersFunctionbuild_backbone

(cfg)

src/models/backbone.py:118

↓ 1 callersFunctionbuild_coco_joint_vis

(cfg, num_cats)

src/datasets/coco_joint_vis.py:133

↓ 1 callersFunctionbuild_criterion

(matcher, num_classes, cfg)

src/models/criterion.py:255

↓ 1 callersFunctionbuild_deformable_transformer

(cfg)

src/models/deformable_transformer.py:359

↓ 1 callersFunctionbuild_devis_ablation_transformer

(cfg)

src/models/devis_ablation_transformer_wo_t_conn.py:110

↓ 1 callersFunctionbuild_devis_transformer

(cfg)

src/models/devis_transformer.py:176

↓ 1 callersFunctionbuild_inference_matcher

(cfg)

src/models/matcher.py:386

↓ 1 callersFunctionbuild_matcher

(cfg)

src/models/matcher.py:215

↓ 1 callersFunctionbuild_position_encoding

(cfg)

src/models/position_encoding.py:127

↓ 1 callersFunctionbuild_tracker

(model, cfg)

src/models/__init__.py:84

↓ 1 callersFunctionbuild_visualizers

(cfg)

src/util/visdom_vis.py:94

← previous next →101–200 of 681, ranked by callers